ES2198388T3 - SYNCHRONIZATION METHOD OF BASE STATIONS. - Google Patents

Abstract

The method involves transmitting synchronization sequence having a primary Golay sequence followed by second Golay sequence obtained from polyphase complementary sequence, from one base station. Another base station effects the correlation of the synchronization sequence with the replica of two sequences by trellis filtering. The correlation result is then added to provide synchronization information.

Description

Station synchronization method base.

The present invention relates to a method of base station synchronization in a system Mobile radio telecommunication More particularly, the present invention relates to a method of synchronizing stations of base for a duplex telecommunication system by time division (TDD). Said telecommunication system is, by example, the ongoing system of normalization, commonly called 3GPP W-CDMA TDD.

A plot of radio of such a telecommunication system. Is made by fifteen time slots of which some, for example the intervals IT_ {o}, IT_ {1}, IT_ {2}, IT_ {5}, IT_ {6} and IT_ {8}, are intended to transport the data (in the broad sense of the term) in a downward direction (base station to mobile terminal) while others, the intervals IT_ {3}, IT_ {4}, IT_ {7}, IT_ {9}, IT_ {10}, IT_ {11}, IT_ {12}, IT_ {13} and IT_ {14}, are intended for data transport upstream (mobile station towards base station). In the course of an interval of transmission, the data (D) is transmitted in the form of a sequence of symbols The interval also includes a midambulo (M) comprising pilot symbols that allow the estimation of channel, a power control word (TPC), and a range of security (GP '). In such a system, several mobile terminals or base stations can send or receive data in the same time interval. The links are differentiated by multiplexing by code division (Code Division Multiple Access = CDMA). The Symbols transmitted by or for different users are spectrally extended, approximately at a frequency `` chip '' 1 / T_ {c} where T_ {c} is the transmission period elementary.

Because the same frequency can be used both upstream and downstream descending, it is imperative to ensure the synchronization of base stations Indeed, if this were not the case, a first mobile terminal that emits with great power in an ascending channel  could interfere with a second mobile terminal, near the first, receiving the data on a descending channel. The synchronization voltage between neighboring base stations is order of a few microseconds (approximately 5) in the system W-CDMA TDD.

To synchronize between stations At the base, several methods have been proposed in the state of technique. According to a first method, synchronization is obtained thanks to GPS receivers that equip base stations. According a second method, first proceed in an initial phase, by example during the installation of the network or a new station base, to a rude synchronization (of the order of a few tens of ms, that is, a few tens of thousands of `` chips ''). Is initial rude synchronization is ensured by the network, more precisely by the radio access controller (RNC) that controls several base stations (also called `` nodes B '') neighbors Then a synchronization is carried out regularly Fine through the radio interface between neighboring base stations. This fine synchronization is primarily intended to correct drifting sequencing clocks between stations neighboring base. For this, certain time intervals are reserved for the transmission and reception of a synchronization signal. A transmission interval dedicated to synchronization comprises essentially a synchronization sequence (sync) and a period of  security (GP). Synchronization is obtained, in a way itself known, by correlation of the sequence received with a Replication sequence of the transmitted. The correlation is made over a temporary window of length given by the margin of precision of gross synchronization. So when a station base receives a synchronization sequence and detects a peak of correlation in this window, it can synchronize its sequencing with that of the surrounding base stations.

The synchronization sequence usually used is long (several thousand `` chips '') in order to get a good correlation accuracy for a power by Acceptable symbol The security period must be longer than propagation time from a base station to a neighboring station in order to avoid, upon receipt, a usurpation of the synchronization sequence over a neighboring time interval. As the distance between two base stations is higher than the radius of a cell, the safety period (GP) is chosen plus larger than the largest normal safety period (GP '). The safety period (GP) must also take into account drift of plot clocks.

The synchronization sequence is chosen to have good autocorrelation properties, namely a peak of Very pronounced autocorrelation. Generally, you get the synchronization sequences used from polynomials primitives on GF (2), Galois body of cardinal 2. Tal sequence has a length L that has a nth power of 2-1, that is L = 2 N -1. This is the case mainly of the Gold called sequences that have been Proposals in report TSGR1 # 15 (00) 0946 entitled `` Sequences for the cell sync burst '' of the Working Group TSG-RAN of the ETSI to synchronize stations neighboring base.

Gold sequences have good properties of periodic autocorrelation. The correlation of a sequence constituted by the repetition of a Gold sequence with a replica of the sequence of the latter does not present peaks Important side However, these sequences do not present unfortunately so good autocorrelation properties aperiódica (correlation of an isolated Gold sequence with a replica). What's more, the correlator used generally operates in the temporal domain in the form of an adapted FIR filter classic that presents a complexity in O (L) that can be very high In addition, the choice of the lengths of such sequences are reduced, since they cannot be taken, as has seen, more than 2 N values -1 and a truncation would lead to a sensitive loss of properties of autocorrelation.

An objective of the present invention is to propose a synchronization method of neighboring base stations thanks to the transmission of a synchronization sequence that presents very Good correlation properties. A secondary objective of the invention is to offer a great choice of sequence lengths possible synchronization, and this for a low degree of complexity  of the correlator.

The present invention is defined by a method. of synchronization of base stations in a system of mobile radio telecommunication, according to which a first station base transmits a synchronization sequence comprising a first sequence followed by a second sequence, forming said first and second sequences a couple of complementary sequences and because at least a second base station correlates of said synchronization sequence with a replica of the first sequence and a replica of the second sequence, being added to then the correlation results to give a signal of correlation that provides synchronization information. The first sequence is separated from the second sequence by a predetermined length separation interval (2GP, 2E), of so that the correlation signal has a correlation peak Isolated in a temporary window.

Advantageously, the first and second sequences they are complementary sequences of Golay.

According to a first embodiment, it is obtained said separation interval providing safety intervals around the first and second sequences.

According to a second embodiment, it is obtained said separation interval providing for a periodic extension of the first sequence followed by a periodic extension of the second sequence

According to a third embodiment, the first sequence by means of a first sequence of Golay and of a first auxiliary sequence successively multiplying said first sequence of Golay by the bits of the first sequence assistant.

Similarly, the second sequence can be generated by means of a second Golay sequence, complementary to said first Golay sequence, and a second auxiliary sequence successively multiplying said second Golay sequence by the bits of the second sequence assistant.

Advantageously, the first auxiliary sequence and the second auxiliary sequence are complementary sequences of Golay

According to a variant embodiment, the correlation by grid filtering.

The characteristics of the invention mentioned above, as well as others, will appear more clearly with the reading the following description made in relation to the attached figures, among which:

Figure 1 schematically represents a emission frame of a transmission system of the type W-CDMA TDD

Figure 2A represents a first mode of embodiment of the invention

Figure 2B represents a second mode of embodiment of the invention

Figure 2C represents a third mode of embodiment of the invention

Figure 3 represents a correlator useful in the third embodiment of the invention.

The general basic idea of the invention is use, for the synchronization of neighboring base stations, a pair of complementary codes, for example complementary codes polyphases, more particularly a pair of complementary codes of Golay Complementary codes, known as such, have as a remarkable property that the sum of its functions of aperiodic autocorrelation is a function of Dirac. Said of another mode, if a pair of such complementary codes is observed (A, B), you have \ varphi_ {AA} (m) + \ varphi_ {BB} (m) = δ (m) where m is the time index, δ the Kronecker symbol and \ varphi autocorrelation function aperiodic

In the continuation of the description it will be done essentially mention of complementary Golay codes. The invention applies however to complementary codes in general.

On the other hand, as described mainly in article S.Z. Budisin, titled `` Efficient press compressor for Golay complementary sequences '' and published in Electronics Letters, Vol 27, nº3, pages 219, 220 in January 1991, the correlator can be made thanks to a filter grid that presents a complexity in O (logL) and not in Or (L) as in a classic FIR adapted filter. This filter of grid is also called EGC filter by Efficient Golay Correlator An example of an EGC filter embodiment is given in the article by B.M. Popovic titled `` Efficient Golay Correlator '', published in IEEE Electronics Letters, Vol 35, nº17, January 1999.

The use of such an EGC filter for the purpose of Synchronization between radio transmitters-receivers is described in WO-A-0054424.

In addition, for a given authorized length, There are several possible Golay sequences. Indeed, like the Golay sequences are generated by generator codes you can show that two different generator codes and the same length generate equally different Golay sequences and of the same length These sequences have great properties of intercorrelation (i.e. low intercorrelation values), allowing use, for example, in groups of base stations, different codes or even synchronize the base stations at different times of your sequencing

A first embodiment of the invention It is illustrated in Figure 2A. According to this embodiment, a synchronization sequence consists of two sequences complementary to Golay A and B multiplexed in time, being preceded and followed each sequence by a security interval, as described in the french application FR-A-9916851 filed on 30.12.1999 on behalf of this same applicant. This sequence is transmitted by a base station and received by a station neighboring base. Upon receipt, the synchronization sequence is correlated with a replica of sequence A and a replica of sequence B, the result of correlation with sequence A is delayed so that it is temporarily aligned with the result correlation with sequence B before they are added, obtaining the Dirac peak when the replicas of A and B are aligned with the corresponding sequences. The presence of GP_ {2} security interval ensures that, during the correlation, sequences A and B do not overlap the replicas corresponding complementary, namely B and A respectively, in a temporary window centered on the alignment position temporary. Thus, the secondary correlation peaks that can result of intercorrelation between sequences and replicas Complementary are rejected outside this window. Plus precisely, if GP_ {2} = 2.GP_ {3} = 2.GP_ {1} = 2.GP, the sum of the two correlation results show an isolated Dirac peak in a window of width 2.GP around the position of temporal alignment The correlations are made advantageously by EGC correlators, such as mentioned above.

A second embodiment of the invention It is illustrated in Figure 2B. According to this embodiment, a synchronization sequence consists of two sequences Golay complementary time multiplexed, being preceded and followed each sequence by a periodic extension, such as It has been explained in the French application entitled `` Séquence d'estimation de canal et procédé de canal d'estimation de transmission utilisant un telle séquence '' presented on behalf of This same applicant. The periodic extension of a sequence given is a truncation of the sequence obtained by repetition of said sequence. To do this, just concatenate the sequence to extend a corresponding prefix at the end and a suffix corresponding to the beginning of said sequence. It has been indicated schematically in to figure 2A the concatenation of prefixes and of suffixes for two complementary sequences of Golay, A and B. The synchronization sequence is in turn made up of two sequences thus extended ext (A) and ext (B). The periodic extensions produce the same advantage as the interval GP_ {2}, namely the absence of correlation peaks side around the Dirac peak in a certain window temporary. More precisely, if the suffixes and prefixes are sized identical and equal to E, the sum of the correlation results it will have an isolated Dirac peak in a window of width 2.E around the temporary alignment position. This is easily understand if you consider the hypothesis in which the synchronization sequence comprises sequences A and B fully periodized. The correlation with replicas of A and B then produces a series of Dirac peaks from period L. One periodic extension of amplitude E is equivalent to truncating this series by a window of width 2.E around the alignment peak temporary. The advantage of this embodiment in relation to precedent is not to cause abrupt signal power variations between sequences A and B, at the level of the transmitter amplifier. These brutal variations can generate high frequencies of intersymbol interference and therefore degrade, upon receipt, The correlation results.

A third embodiment of the invention It is illustrated in Figure 2C. According to this embodiment, generates, from a sequence of Golay A or B and a sequence auxiliary X, a composite sequence, according to the construction mode of the hierarchical sequences. More precisely, it multiplies successively the first bit of the auxiliary sequence X for all bits of sequence A, then the second bit of the second sequence by all bits of sequence A, so on and the sequences obtained are concatenated. We will point out in the following by A * X such a composite sequence, with A being the base sequence and X the auxiliary generatrix sequence. Golay sequences complementary A and B can thus be multiplexed by sequences identical and different auxiliary X and Y, which may also be latest also Golay sequences.

Let A * X and B * X be composed sequences obtained from a pair A, B of complementary Golay sequences, of length L, extended by prefixes or suffixes of size E. A * X and B * X are multiplexed in time and separated by an interval W. The received signal is correlated with the sequence A of one party and with the sequence B of another party. The result of the first correlation is delayed by (L + 2E) + W and is added to the result of the second correlation. The sum obtained is a sequence R comprising a series of Dirac peaks of period L '= L + 2E modulated by the values x_ {o}, x_ {1} .... X_ {k} where K is the length of the sequence x, surrounded each peak by a window of width 2.E that contains only zeroes. The sequence R is then subjected to filtering by a linear response filter H (z) = x_ {0} + x_ {1} .z ^ {- L '} + ... + x_ {k} .z ^ {- K.L'}

The filtered R sequence contains a Dirac peak of height 2.K.L in the center of a null window of width 2.E lo which allows to detect it easily. In addition, the total sequence consisting of A * X and B * X sequences multiplexed in time is of total length 2. (L + 2.E) .K + W, which offers a great choice of authorized sequence lengths.

According to another embodiment variant, it is generated four sequences composed A * X, A * Y, B * X, B * Y where A, B form a first pair of extended Golay complementary sequences or no and X, Y form a second pair of complementary sequences of Golay that serve as auxiliary generating sequences.

Composite sequences are multiplexed in the time and separated by intervals that we will assume equal and of width W. The sequences A and B are of length L '= L + 2.E where L is the length of the base sequence and E the size of the extension, being the sequences X, Y of length K. The length of the total sequence is therefore 4 (L + 2E) K + 3W, which offers a great choice of sequence lengths authorized.

This variant takes advantage of the fact that L 'pairs of complementary sequences (X, Y) are available under the form of subsequences S_ {m} and S_ {m} 'with S_ {m} (n) = (A * X) _ {n.L '+ m} and S_ {m} '(n) = (B * X) _LL + m}, m = 0, ...., L'-1 obtained by decimation of the total sequence initial. Instead of making a correlation with a EGC correlator, a `` hierarchical '' correlator is used, the first stage of the modified EGC function correlator such as depicted in figure 3.

Assume that the pair of X and Y sequences has been classically generated by an elementary sequence s_ {0}, ..., s_ {K-1}, where K = 2 k -1 and delays D '0 , D '1, ..., D' K-1 with D 'i = 2 Pi where (P 0, P 1, ..., P_ {k-1}) is a permutation on the set (0, 1, ..., k-1) in the following recursive way: X_ {0} (i) = δ (i); Y_ {o} (i) = \ delta (i); \ belowdisplayskip = .5 \ baselineskip X_ {n} (i) = X_ {n-1} (i) + S_ {n-1} .X_ {n-1} (i-D'i); \ belowdisplayskip = .5 \ baselineskip Y_ {n} (i) -S_ {n-1} .Y_ {n-1} (Di-D 'i);

Similarly, suppose the pair of sequences A, B has been generated by the elementary sequence t_ {0}, ..., t_ {l-1}, where L = 2 ^ -1 and delays D_ {0}, D_ {1}, ..., D_ {K-1} with D_ {i} = 2 PI where (P_ {0}, P_ {1} , ...., P_ {l-1}) is a permutation on the set (0, 1, ..., l-1).

The first correlation stage performs a correlation with the pair of sequences X, Y, but differs from a classic EGC correlator because the delays have been multiplied by a factor L 'to take into account the dispersion of the samples. The two correlation results are added after of the time alignment by a delay D_ {XY} the delay D_ {XY } which separates the A * X and A * Y sequences from one part, and the B * X and B * Y sequences, on the other hand. The second stage of correlator correlates with sequence pair A, B And it is itself classic. The correlation results are aligned temporarily for a delay D_ {AB} and added, corresponding the delay D_ {AB} to the temporal difference between the sequences A * X and B * X, on the one hand, and the sequences A * Y and B * Y, on the other part.

The correlator so constituted operates first a gross correlation with a step L 'and then a Fine correlation to the sampling step. Its complexity is low since that the number of operations carried out is in O (log (K) + log (L)).

Although the example described above only it has two levels of sequences and two levels of correlation, the invention is immediately generalized to any number of sequence levels and corresponding stages of the hierarchical correlator.

Claims (9)

1. Synchronization method of base stations in a mobile radio telecommunication system, according to which a first base station transmits a synchronization sequence comprising a first sequence followed by a second sequence, said first and second sequences forming a pair of complementary sequences and at least a second base station effects the correlations of said synchronization sequence with a replica of the first sequence and a replica of the second sequence, the correlation results are then added to give a correlation signal that provides a synchronization information, characterized in that the first sequence is separated from the second sequence by a predetermined length separation interval (2GP, 2E), so that the correlation signal has an isolated correlation peak in a time window and that the secondary correlation peaks that may result of the intercorrelation between sequences and complementary replicas are rejected outside said time window. 2. Synchronization method according to claim 1, characterized in that the first and second sequences are complementary Golay sequences. 3. Synchronization method according to claim 1 or 2, characterized in that said separation interval is obtained by providing safety intervals around the first and second sequences. 4. Synchronization method according to claim 1 or 2, characterized in that said separation interval is obtained by providing for a periodic extension of the first sequence followed by a periodic extension of the second sequence. 5. Synchronization method according to claim 2, characterized in that the first sequence is generated by means of a first Golay sequence and a first auxiliary sequence by successively multiplying said first Golay sequence by the bits of the first auxiliary sequence. 6. Synchronization method according to claim 5, characterized in that the second sequence is generated by means of a second Golay sequence, complementary to said first Golay sequence, and a second auxiliary sequence by successively multiplying said second Golay sequence by the bits of the second auxiliary sequence. 7. Synchronization method according to claim 6, characterized in that the first auxiliary sequence and the second auxiliary sequence are complementary Golay sequences. 8. Synchronization method according to one of the preceding claims, characterized in that the correlation is carried out by grid filtration. 9. Synchronization method according to one of the preceding claims, characterized in that said time window is of length equal to said predetermined length and does not contain secondary correlation peaks.